From 04cfce2eb845c0dfc7d578a6fb3c44cb546b2028 Mon Sep 17 00:00:00 2001
From: Bnyro <bnyro@tutanota.com>
Date: Mon, 16 Oct 2023 10:45:09 +0200
Subject: [PATCH] [feat] yep: support for images and news (and safesearch)

---
 searx/engines/yep.py | 79 ++++++++++++++++++++++++++++++++++++++++++++
 searx/settings.yml   | 30 +++++++++--------
 2 files changed, 95 insertions(+), 14 deletions(-)
 create mode 100644 searx/engines/yep.py

diff --git a/searx/engines/yep.py b/searx/engines/yep.py
new file mode 100644
index 000000000..c3cb65c7b
--- /dev/null
+++ b/searx/engines/yep.py
@@ -0,0 +1,79 @@
+# SPDX-License-Identifier: AGPL-3.0-or-later
+# lint: pylint
+"""Yep (general, images, news)
+"""
+
+from datetime import datetime
+from urllib.parse import urlencode
+from searx.utils import html_to_text
+
+about = {
+    'website': 'https://yep.com/',
+    'official_api_documentation': 'https://docs.developer.yelp.com',
+    'use_official_api': False,
+    'require_api_key': False,
+    'results': 'JSON',
+}
+
+base_url = "https://api.yep.com"
+search_type = "web"  # 'web', 'images', 'news'
+
+safesearch = True
+safesearch_map = {0: 'off', 1: 'moderate', 2: 'strict'}
+
+
+def request(query, params):
+    args = {
+        'client': 'web',
+        'no_correct': 'false',
+        'q': query,
+        'safeSearch': safesearch_map[params['safesearch']],
+        'type': search_type,
+    }
+    params['url'] = f"{base_url}/fs/2/search?{urlencode(args)}"
+    params['headers']['Referer'] = 'https://yep.com/'
+    return params
+
+
+def _web_result(result):
+    return {
+        'url': result['url'],
+        'title': result['title'],
+        'content': html_to_text(result['snippet']),
+    }
+
+
+def _images_result(result):
+    return {
+        'template': 'images.html',
+        'url': result['host_page'],
+        'title': result.get('title', ''),
+        'content': '',
+        'img_src': result['image_id'],
+        'thumbnail_src': result['src'],
+    }
+
+
+def _news_result(result):
+    return {
+        'url': result['url'],
+        'title': result['title'],
+        'content': html_to_text(result['snippet']),
+        'publishedDate': datetime.strptime(result['first_seen'][:19], '%Y-%m-%dT%H:%M:%S'),
+    }
+
+
+def response(resp):
+    results = []
+
+    for result in resp.json()[1]['results']:
+        if search_type == "web":
+            results.append(_web_result(result))
+        elif search_type == "images":
+            results.append(_images_result(result))
+        elif search_type == "news":
+            results.append(_news_result(result))
+        else:
+            raise ValueError(f"Unsupported yep search type: {search_type}")
+
+    return results
diff --git a/searx/settings.yml b/searx/settings.yml
index be420528f..1e9f0855b 100644
--- a/searx/settings.yml
+++ b/searx/settings.yml
@@ -484,23 +484,25 @@ engines:
       website: https://crowdview.ai/
 
   - name: yep
-    engine: json_engine
+    engine: yep
     shortcut: yep
     categories: general
+    search_type: web
+    disabled: true
+
+  - name: yep images
+    engine: yep
+    shortcut: yepi
+    categories: images
+    search_type: images
+    disabled: true
+
+  - name: yep news
+    engine: yep
+    shortcut: yepn
+    categories: news
+    search_type: news
     disabled: true
-    paging: false
-    content_html_to_text: true
-    title_html_to_text: true
-    search_url: https://api.yep.com/fs/1/?type=web&q={query}&no_correct=false&limit=100
-    results_query: 1/results
-    title_query: title
-    url_query: url
-    content_query: snippet
-    about:
-      website: https://yep.com
-      use_official_api: false
-      require_api_key: false
-      results: JSON
 
   - name: curlie
     engine: xpath